This story examines how rising unreported non-response in the Australian Labour Force Survey (LFS), is forming a burgeoning lumpenproletariat, because it hides the real unemployment figures. The ABS unemployment statistics reported by a lazy media, have given comfort to big business and government, while ignoring a significant threat to the Australian economy and society.

According to the ABS, Australia’s unemployment figures were now better than before Covid-19 hit. The official ABS data show the unemployment rate dropping from 4.6 to 4.2 per cent, with an estimated 64,800 jobs created between November and December. These figures are laughable.

Sydney and Melbourne CBD’s are wastelands (Adelaide was bad before the virus). Hospitality is on its knees, international air travel is comatose and our universities are tumbleweed cities. People have dropped out of the workforce in droves.

There are about 13.8 million people in labour force. 26,000 specific households (about .32 per cent of the civilian population over 15 years of age) are sent a survey each month for eight months, with one-eighth of the sample replaced each month to bring in new people.

The first interview is completed face-to-face and subsequent interviews are conducted by phone and online. The ABS preference, for cost reasons, is for people to reply online. The official LFS survey non-response rate is between 5-7 per cent.

There has been considerable discussion about the tight definition of unemployment: “all persons aged 15 years and over who were not employed during the reference week and had actively looked for work and were available to work or were waiting to start a new job.” If they worked one hour or more in the reference week, they are employed.

All OECD nations including Australia, use the same methodology and roughly the same definitions.

We are witnessing a staggering rise in the number of discouraged ‘job seekers’, officially around 300,000, although no one really knows the exact figure. I suggest the true figure is closer to 700,000. Since the GFC and before, they’ve been knocked back from jobs either because of their age, lack of schooling, training, experience or age or race prejudice. Some were the victims of the GFC, previous recessions and more recently, the Covid-19 lockdowns and retail stagnation.

These discouraged job seekers are members of a rising Australian lumpenproletariat or civil dead.

Structural economic change is destroying social capital. Elections are held, governments are formed but the system fails to deliver. It’s no wonder eligible voter numbers are falling but at the same time, LFS non-response is rising.

Some years ago I worked for the Australian Government in Labour Market Strategy in Canberra. I examined the demographic profiles of some of the poorest regions in Australia (once called the Priority Employment Areas). The Federal Government, its departments and agencies, are generally not held in high regard by the unemployed in these areas. The longer one is unemployed, the greater the antipathy and ennui.

Those who were hounded by Robodebt or so-called ‘job placement’ agencies, don’t give a damn for labour force data or those collecting it. They do not want to answer questions which reaffirms their unemployed status. When one is battling poverty, the requirement to complete and lodge eight Labour Force Surveys, whether online or by telephone, ranks very low on their hierarchy of needs.

Those who do not respond to the survey or who fail to complete each survey, are excluded from the count. They are ‘adjusted’ later in the weighting process. Weighting is a mathematical technique that makes the results reflect variables such as non-responders. Data from previous local population surveys is used to ‘fill in’ the missing information. The data is massaged to show validity rather than accuracy. Non-responders or incomplete responders are mathematically manipulated out of the count. Statistical validity is not the same as representative truth.

The ABS state that between 93 and 95 per percent of people complete the LFS survey. There’s a legal compulsion to complete the survey but it’s rarely used. This is an extraordinarily low non-response rate compared to other nations using the same methodology. So low, it’s BS.

In the UK, LFS non-response rates fell to 20 per cent in the early 1990s and are now around 40 per cent. Those who dropped out were mainly in the 20-29 years age group, unemployed and in large households.

In the USA, non-response rates are around 12-15 per cent and sky rocketed to 35 per cent during the worst of the Covid-19 fallout in March and May of 2020. Non-response in some of its largest national surveys is climbing because poorer Americans have lost faith in their political system.

American studies show that Blacks, Hispanics, the unemployed and those subsisting on the margins of society, tend to not complete the American version of the LFS, which leads to undercounts of up to 30 percent for those cohorts. That alone accounts for a 2 per cent underestimate in the national unemployment figures.

The Australian media doesn’t interrogate the unemployment figures. Its fetish for numbers coincides with a decline in an understanding – or rising apathy – of how those statistics are created and what they mean. Statistics are a shadow of the phenomena they claim to represent, not the thing itself. We ascribe a veracity to them which is dangerous and foolhardy.

Since the mid 1980s, the labour market has been undergoing radical change with the atomisation of various industries, the decline in fulltime jobs, the causualisation of the workforce, the rise of the ‘gig economy’ and ‘robot jobs’, people dropping out of the workforce and the fallout from the GFC. There will also be significant long terms effects in the labour market from the Coronavirus.

The unemployment figures are massaged and weighed through a statistical machine that feeds delusion rather than clarity. We need tools and methods that accurately reflect the dynamics of a rapidly changing labour market and the rise of a large, disaffected underclass.